Project 3 Summary ? Abstract Recent animal work from our laboratories suggests that the synapses between hair cells and cochlear nerve terminals are the most vulnerable elements of the inner ear in both noise-induced and age-related hearing loss. This synaptic degeneration does not affect hair cells or thresholds and, therefore, hides behind a normal audiogram. However, this ?hidden hearing loss? likely contributes to decreased ability to understand speech, especially in noisy environments. It may also be an underlying cause of tinnitus and hyperacusis. In animals, hidden hearing loss can be diagnosed by measuring the suprathreshold amplitude of ABR wave I, the sound-evoked activity of the cochlear nerve. We have begun translating these insights from animal work to human studies. We have a pilot study from audiometrically normal young adults showing significant correlations between electrocochleographic measures consistent with cochlear synaptopathy and performance on difficult speech recognition tasks. Here we propose to extend our pilot study into large-scale cross-sectional (Aim 1) and longitudinal (Aim 2) studies of hidden hearing loss in college students with normal audiometric thresholds and widely differing lifestyle with respect to aural abuse, as quantified by questionnaire. For Aim 1, we will test the hypotheses that hidden hearing loss, defined as performance deficits on difficult word- recognition tasks, is correlated with physiologic response deficits consistent with cochlear synaptopathy and with estimated lifetime noise dose. Our outcome measures will be speech-in-noise tests. A statistical model will test the correlation of these outcomes with noise-exposure history and with a battery of physiological or psychophysical measures chosen to probe different stages of auditory processing, i.e. distortion product otoacoustic emissions and high-frequency audiometry, SP/AP ratio and envelope following responses to tones at high modulation frequencies, middle ear muscle or medial olivocochlear reflexes, several variants of frequency following response probing monaural and binaural temporal fine-structure processing, a temporal integration test of theories on stochastic undersampling in cochlear synaptopathy, and tinnitus severity/handicap and loudness discomfort level/hyperacusis. Using principal components analysis, cluster analysis and adaptive LASSO, we will find the test combination that best predicts the outcome measures and assess the relative contributions of peripheral vs. central pathophysiology to the observed performance deficits. In Aim 2, we will test the hypothesis that hidden hearing loss progresses in young adults with regular and continued acoustic overexposure by tracking a cohort of students over the five-year period of this project using the same test battery as described in Aim 2.